Energy management system for PV-based distributed generators in AC microgrids using an adapted JAYA optimizer to minimize operational costs, energy losses, and CO2 emissions

The integration of photovoltaic (PV) distributed generators into alternating current (AC) microgrids poses significant challenges due to the intermittency of solar generation and the need to maintain the dynamic balance between generation and demand. Additionally, energy losses, nodal voltage and line current constraints, and generator capacity limits further complicate the efficient planning of these networks. This article addresses these challenges by developing a mathematical model aimed at minimizing operational costs, reducing energy losses, and decreasing C O 2 emissions. The model incorporates key constraints of microgrids in distributed generation environments, such as power balance, generation limits, nodal voltages, and line currents. To solve the model, an adapted version of the JAYA optimization algorithm was implemented, and its performance was compared against four established methodologies: the Chu & Beasley Genetic Algorithm (CBGA), Particle Swarm Optimization (PSO), the Vortex Search Algorithm (VSA), and Ant Lion Optimization (ALO). The analysis was conducted using two test microgrids: one with 33 nodes operating in grid-connected mode and another with 27 nodes operating in islanded mode, utilizing generation and demand data from Medellín and Capurganá, Colombia. Simulations were performed 100 times per scenario to evaluate the optimal solution, average performance, repeatability, and processing times. The results demonstrated that the adapted JAYA algorithm was the most effective for both test systems. Achieving average reductions for both scenarios of 31.23%, 40.61%, and 31.53% in operational costs, energy losses, and C O 2 emissions, respectively. • Mathematical model for energy management of distributed generation in grid-connected and isolated AC microgrids. • Master-slave methodology with the JAYA algorithm to optimize technical, economic, and environmental performance. • Statistical analysis validating the robustness and effectiveness of the methodology under varying operational scenarios. • Enhanced benchmark results achieved for 27-node and 33-node test systems through advanced optimization techniques.